Novel methodology for turbine gas meters error curve modelling across a wide range of operating parameters

Abstract

Abstract In this paper, performance of turbine flowmeters was investigated for different flowmeter ranges and working gas operating pressures. Variation of these parameters was represented in dimensionless form as a function of Reynolds Number and gas density ratio, while the relative flow measurement error was selected as the most important operating characteristic. A novel error curve model based on turbine machine theory and dimensionless analysis was introduced for the purpose of error data fitting across a wide range of gas flow rates and operating pressures. The main advantage of the presented model is the capability of accurate error data fitting with a single continuous equation, as demonstrated by high R 2 {R^{2}} values for the vast majority of flowmeters analyzed in this study. The acceptability criterion was designed based on the fact that the expanded measurement uncertainty of the relative error must not exceed 0.5 %. Besides an accurate interpolation, our model can also be utilized for prediction of turbine flowmeter performance at modified flow conditions (pressure and flow rate, working gas properties), and for assessment of the drift of flowmeter performance over time. The novel error curve model is demonstrated to outperform the standard polynomial-based model regardless of the independent variable used.